Funds are requested for a 5-year continuation of a project examining the postseparation adjustment of mothers and their children (R01-MH38318). Based on work already carried out in the passive longitudinal study, the proposed investigation implements a theory-driven intervention designed to prevent adjustment problems in families in the divorce process. The proposed randomized experimental longitudinal design provides for systematic intervention on factors previously found to relate to the initiation and maintenance of adjustment problems. A new sample of 220 recently separated mothers and boys (Grades K-2) and 60 female siblings (Grades K-6) will be recruited. Families will be randomly assigned to control or experimental conditions. Multimethod/agent assessment will be conducted at baseline, termination, two follow-up probes at 6-month intervals, and a third and final follow-up assessment 12 months later. One objective is to experimentally manipulate variables hypothesized to influence child adjustment (e.g., development of antisocial behavior patterns) and maternal adjustment problems (e.g., high levels of stress and depression) in divorce. Mothers in the experimental group will receive an intervention designed to prevent and ameliorate adjustment problems for herself and her children by teaching skills hypothesized to control the problems. Measures of skills will then be examined for the independent and interacting contributions to the course of subsequent child and maternal adjustment. A second objective is to test the outcome effects of the intervention. A third objective is to study the differential postseparation adjustment of boys and girls. A fourth objective is to replicate and extend theoretical models of child and maternal adjustment evaluated with the first sample. The data sets will enable us to test process models examining mechanisms hypothesized to explain why some mothers and children have long-term adjustment problems following separation and others do not.